1.1 Rationale for the Study

The rationale for our study was to test the influence of cancer awareness, age, and gender on screening habits for a diverse pool of cancers. Previous studies have focused on the most well documented cancers, and this has created a gap for evidence to drive studies and programs about other cancers, specifically, colon, liver, ENT and, gastric cancers. Evidence from the study can improve cancer awareness and screening promotion campaigns by addressing a broad range of cancers, which are all a burden to the Ghanaian health system.

1.2 Objectives

1.3 Theory

The Health Belief Model (HBM) is used as the theoretical framework for this study. The HBM, a social-psychological theory from a social cognitive perspective, was developed in the United States in the 1950s to help predict preventive health behavior. The model was used by Hochbaum [17] in a 1956 study of X-Ray screening for Tuberculosis. At the time of Hochbaum's study, there was evidence that some TB patients could be asymptomatic and thus not show any outward signs of infection. Screening was promoted as a safe way to detect infection rates early and reduce its spread amongst the population. In Hochbaum [17], perceived susceptibility of the patient to TB and the perception that screening is beneficial, predicted their participation in screening programs. Rosenstock [18] noted that several studies followed with the aim of using the HBM ideas to increase perceptions of susceptibility and improve awareness about the benefits of screening. In recent years, self-efficacy has been included in the model to account for task specific self-confidence [19].

Key components of the HBM include perceived susceptibility, perceived severity, health motivation, perceived benefits, and perceived barriers [20]. These perceptions and motivation are influenced by demographic and psychological variables [19]. The action taken by the individual is predicted in a manner reflecting their perceptions and motivation. Susceptibility and severity deal with perceptions of risk and the evaluation of seriousness associated with a consequence. Health motivation refers to the concern an individual demonstrates about healthcare topics. The perceived benefits in the HBM relate to the subjective idea that an action like cancer screening reduces severity or threats. The perceived barriers mentioned in the model focus on disadvantages of acting. The “desirable” behavior to be predicted by all these perceptions and motivation is termed “action” which is influenced by cognitive prompts or cues to act to increase the perceived benefits. These can range from regular screening to regular visits to a health professional.

Abraham and Sheeran [20] reported diverse applicability of the HBM to diseases, including hypertension control, alcohol, smoking, vaccination, diabetes, and psychiatry. Taking a cue from this diverse applicability of the HBM and previous studies such as Belay et al. [21], Nyaaba and Akurugu [22], the current study used age, gender, and awareness to predict cancer screening.

2.1 Study Design

This study employed an online cross-sectional survey. The questionnaire presented items to measure the cancer awareness, gender, age, and screening behavior of participants.

2.2 Study Setting

The online survey was conducted in several Ghanaian cities namely, Accra, Tema, Kumasi, Cape Coast, Ho, Tamale, Takoradi, and Sunyani. These are regional capitals and are the most affluent cities in Ghana with a high rate of internet penetration and active cancer outreach campaigns. Based on projections by the Ghana Statistical Service, the estimated adult literate population of these cities is 6000 000 [23]. The regional capital status of these cities also provides them with Regional Hospitals capable of screening for most of the common cancers identified in Ghana. Additionally, there are numerous private health facilities with cancer screening services in these cities. Cancer awareness is higher in these “metropolitan” areas because of active NGOs and government campaigns aiming to increase awareness and screening. Survey links were sent by the researchers to the participants between June and August 2021. To prevent individuals from answering twice, they received the links once and after completion they informed the individual who recruited them. These recruiters did not reveal the identities of the respondents to the researchers.

2.3 Participants

To be included in this study, participants were required to be adult, have internet access and be resident in Accra, Tema, Kumasi, Cape Coast, Ho, Tamale, Takoradi or Sunyani, all in Ghana. After meeting these criteria, the online questionnaire presented a question asking the individuals whether they were interested in completing the survey online before links were sent to them. The first batch of 100 participants were randomly recruited from the contacts of the researchers and universities in the above-named cities. Subsequently the snowball technique was used to recruit 407 more participants.

2.4 Variables

The outcome variable in this study was cancer-screening behavior. Cancer awareness, gender, and age were predictor variables. Cancer screening behavior was measured as a dichotomous variable with participants indicating whether they had a history of cancer screening. To measure cancer awareness, participants were asked whether they knew about breast cervical, colon, stomach, liver, prostate and ear, nose, and throat or ENT cancers. Gender was a dichotomous variable, and participants could only select either female or male. Age was measured on a continuous scale with participants stating how old they were at the time of the study. These variables were used to predict cancer-screening habits because institutions such as the Ghana Health Service, Ghana's Ministry of Health, and NGOs organize campaigns to create awareness of these cancers. These campaigns target specific genders for some cancers and encourage older people to be screened. This observation and support from research make these three variables important in the prediction of screening habits.

2.5 Data Sources/Measurements

Data collection: An online survey was deployed on google forms. All data were downloaded and analyzed with SPSS.

Data collection tool: The questionnaire was in three sections. These were cancer awareness, screening history and demographics. The demographics sections collected information about age and gender while the screening history section presented two items which asked participants “Have you ever screened for any cancer?” and “If you have screened for cancer, which type?”. To measure awareness, seven cancers, namely, breast, cervical, colon, stomach, liver, prostate and ear, nose, and throat or ENT were listed. Participants were asked “Have you ever heard of the following?”. The use of single item measures for some key variables such as awareness and screening habit is supported by previous studies such as Williams [24] and Bowling [25].

2.6 Bias

To avoid the confounding effects of professional cancer knowledge from medical professionals, the research widely distributed the questionnaire to non-medical professionals.

2.7 Sample Size

The study size was derived from a random sampling of adults who were on various WhatsApp groups of the researchers and their contacts. The sample size was computed based on the estimated population of adults who were literate and had access to the internet in the cities covered by the study. Using the estimated population of 6000 000, a 95% confidence interval and a 5% margin of error, the required sample size was 385. However, due to the researcher's data collection efforts and the resounding interest of the target population in the study, 507 responses were recorded. The target population of literate adult internet users was chosen because of the nature of current cancer awareness campaigns and observations. Current campaigns mostly use English and deploy campaigns through multiple channels including the internet and offices. The researchers have additionally observed from clinical practice that, most patients who request screening services are literate and exposed to cancer awareness messages.

2.8 Quantitative Variables and Statistical Methods

This statistical approach for this study was quantitative. Both univariate and multivariate analysis were conducted. To ensure consistency and data quality only complete responses were included in the final dataset. This data set was analyzed with version 26 of IBMs SPSS. Univariate analysis (see Table 1) presented the frequencies and percentages of variables. To ensure that all variables to be included in the data were not perfectly correlated and to reveal the strength of relationships, a spearman rho correlation analyses (see Table 2) was conducted. This computation included age, gender, cancer awareness, and screening habit. A model which tested the hypothesis that cancer awareness, age, and gender predict cancer-screening habits was tested with a logistic regression (see Table 3).

2.9 Ethical Considerations

The study adhered to the ethical principles of anonymity, confidentiality, and informed consent. To maintain anonymity, personal identifying information such as names, email, phone numbers, and signatures were not collected. Researchers assured participants that all responses were confidential and there was no way to trace their responses to them. Additionally, data from the study was kept on secure cloud servers accessible only to the researchers. Participants were required to agree to the informed consent statement before they could proceed with participation. To maintain anonymity, participants were not required to sign or type their names on the statement. Any participant who disagreed with the informed consent had the option to refuse participation in the study. Participation therefore meant the participant has agreed to the informed consent statement. Ethical approval for this project was granted by the CCTH IRB with a Federal wide Assurance Number (FWA) of IRB00014450 and a reference number of CCTHERC/EC/2024/189.

3.1 Participants

A total of 507 responses were received but only 503 were included in the final statistical analysis. Four responses were excluded because they were incomplete. The participants were aged between 18 and 74 with a mean age of 32.74 and a standard deviation of 9.10. The responses were received from 309 females, which accounted for 61.4% of responses and 194 males, accounting for 38.6%.

3.2 Descriptive Data

A breakdown of the data on age showed that 48.6% (244) of respondents were aged 19–30, 31.5% (158) were 31–40, and 19.9% (100) were 41–74 years old. Responses indicated that most of the participants were aware of the cancers listed in our research but had never been screened for any of them (see Table 1).

3.3 Outcome Data

In the two-stage multivariate analysis, a correlation treated all variables the same and a logistic regression used cancer-screening habits as the outcome. The first stage tested for perfect correlations and the relationship between age, cancer awareness, cancer screening habits, and gender. In the second stage, cancer-screening habits was used as an outcome, in line with the objectives of this study.

3.4 Main Results

3.5 Summary of Results

In summary, the descriptive results point to a high level of awareness and low screening uptake. Respondents were aged between 19 and 74 (mean = 32.74, standard deviation = 9.10). The age range was appropriate because of the list of cancers considered. Gender and age predicted screening habits. Results indicated that awareness did not predict screening habits. These results are discussed in the next section.

4.1 Key Results

Discussing the hypotheses, the first notable finding is the support for the role of gender in predicting screening habits. In this study, females reported more active screening habits than males. This may be because of the relatively consistent campaign to convince females to screen for breast and cervical cancers [4]. Entire months of the year are dedicated to breast and cervical cancer awareness programs. Again, there are many advocacy groups run by female survivors of these cancers that create awareness and encourage screening. With the possibility of self-screening for human papilloma virus, screening for cervical cancer has invariably become patient centered and thus improved compliance and uptake of cervical cancer screening. These in addition to the WHO's call for elimination of cervical cancer may have contributed to the increased awareness and higher screening practices among women. Comparatively, campaigns for men have focused on Prostate cancer screening from the age of 40 and this makes age an influential factor in the habits of men. Unfortunately, though liver cancer is a leading cause of cancer death and is associated with known causes of chronic liver disease such as hepatitis B and C virus infection, chronic alcohol abuse and aflatoxin infection, only few campaigns are ongoing about the prevention and early detection of liver cancer. The same scenario plays out for gastrointestinal cancers (stomach and colon).

In the second finding, age predicted screening habits. Campaigns for cancer screening have signaled an increased risk with age especially for prostate cancer. This may have led to increased likelihood of screening as one gets older, especially in males. It is thus not a surprising finding. The results suggest that the likelihood of screening increases with age and older people are more likely to screen for multiple cancers. As mentioned by some studies such as Shirazu et al. [7], most cancer screening campaigns in Ghana advocate screening for older people. For instance, for prostate cancer screening, men are advised to screen after the age of 50 while women are advised to start screening after age 25 for cervical cancer. Similarly, for breast cancer, women are mostly advised to screen later in their 40s.

The third finding, which suggested that cancer awareness, did not predict the likelihood of screening, was surprising. The finding may however be explained by several barriers which can impede screening as identified by Ampofo et al. [9] and Ebu et al. [6]. While individuals may be aware of the need to screen many face challenges accessing screening centers and funding the procedure. This maybe a hurdle which many have not been able to surmount. Again, many countries where screening has been successful run a central call and reminder system that ensures that patients are reminded of screening schedules and get scheduled for screening at a health facility near them. Furthermore, in those countries, the cost of screening and treatment for detected lesions is borne by the state, ensuring that beneficiaries do not have to worry about cost. The absence of these factors may negatively impact the screening habits of our respondents.

In summary, findings suggest that age and gender predict screening behavior better than awareness. The researchers can speculate that the reason for the lack of a relationship between awareness and screening could be barriers such as unavailability of affordable screening centers and the absence of an organized screening reminder system.

4.2 Limitations

Despite the predictions by age and gender, the level of screening was still lower than 40% for the respondents. This could be attributed to the fact that a majority of our respondents are younger than 40 years although it was anticipated that this trend would be balanced by a higher representation of women given that cervical cancer and breast screening typically start almost a decade earlier than age 40. Therefore, the findings can only be generalized to adults around the mean age of 32. Additionally, the recommendations about increasing access to screening may be too general because the study focused on screening behavior and not barriers faced by respondents.

Further studies with larger and less skewed population may reveal the peculiarities of screening habits in different age groups.

4.3 Interpretation

Despite several publicity campaigns, cancer screening has not reached the numbers desired by healthcare stakeholders [9]. In the current paper, screening habits were low, though cancer awareness was relatively high. The increasing risk and disease burden posed by other cancers like colon, ENT, gastric, and liver cancer needs to be dealt with by the healthcare system. It was thus positive to note that our respondents were becoming more aware of the diverse nature of cancer and its ability to occur in other parts of the human body. Studies, such as that by O'Mahony et al. [15], have supported the important role of age and gender in predicting screening habits, a suggestion that is also upheld by findings in this study. These also conform to suggestions by the HBM [17].

Findings from this study have two clinical implications. First, for public health practice, findings imply that cancer awareness messages must be improved to include messages about screening based on age and gender. This is with regards to its importance and where screening centers can be found. Second, for clinicians offering screening services, findings imply that these services should be publicized, and practitioners should collaborate with public health officials to improve the visibility.

4.4 Generalizability

This study included most of the commonly diagnosed cancers in Ghana. These cancers exert a significant toll on both personal and public health budgets in Ghana and other African countries. Participants were selected from the biggest cities in Ghana where prevention and early detection campaigns are more frequent, but morbidity and mortality continue to rise. The scope of this study permits an extension of findings to cancer screening campaigns for both genders in cities where screening facilities are available. These include both private and public screening centers as well as periodic screening events, which run for months and are dedicated to specific cancers such as the breast cancer. Findings can help practitioners understand the scope of awareness, its impact on screening and the roles played by both gender and age. Additionally, the findings can help practitioners and researchers understand the current gap between cancer awareness campaigns and screening habits.

4.5 Conclusion

Age and gender play important roles in predicting screening habits in relation to the broader scope of cancers sampled. Older individuals and women are more likely to screen for cancer.

Awareness did not predict screening habits.

The study extends suggestions that screening needs to be a core part of campaigns by healthcare interest groups. We recommend that efforts that improve the ease and likelihood of screening should be intensified.